Differential Regulation of Vascular Endothelial Growth Factor (VEGF-A) Secretion by Insulin and Insulin-Like Growth Factors (IGF) in Human Luteinizing Granulosa Cells (LGC) Following Controlled Ovarian Stimulation (COS) in a Polycystic Ovarian (PCOS) Population

Abstract

To investigate the role of IGF-I and -II in the physiological regulation of VEGF-A production by LGC from preovulatory follicles in PCOS patients in the presence and absence of LH. A prospective, comparative, experimental study In an IRB approved protocol, LGC were collected via follicular aspiration from women (n=6) with PCOS undergoing COS for in-vitro fertilization. The diagnosis of PCOS was established using the ASRM/ESHRE consensus guidelines. The mean age of the study population was 30.2 ± 0.82 years. LGC were isolated and plated at 50,000 cells/well on fibronectin in DMEM-Ham’s F-12 medium with 5 mu;g/ml transferrin, 5 ng/ml selenium, 10 μg/ml aprotinin, 25 mu;g human LDL. Cells were incubated with 0, 1, 10, and 100 ng/ml of r-h IGF-1, IGF-II, and insulin (Sigma) in the presence or absence (control) of 100 ng/ml LH (Ares Serono). LGC were cultured for 72 hours and media were collected daily and assayed for free VEG-F (Quantikine VEGF ELISA, R&D Systems) and progesterone (P). After 3 days, wells were fixed with a Crystal Violet stain and analyzed for DNA content to estimate cell numbers. Data were normalized to DNA content and analyzed using a linear regression model, a mixed effects model, and adjusted using the Bonferroni method for multiple comparisons Exposure to IGF-I, IGF-II and insulin typically increased (p<0.05) DNA content in cell cultures, necessitating the normalization of P and VEGF levels to DNA content. The presence of 1-100 ng/ml IGFs or insulin did not alter control or LH-stimulated P levels during 3 days of culture. Under control conditions, LGCs secreted detectable levels of VEGF-A throughout culture. Addition of IGF-I, IGF-II or insulin (1-100ng/ml) caused a dose-dependent increase (p<0.05) in VEGF levels by day 2-3 of culture. In the absence of LH, physiological levels (1-10 ng/ml) of IGF-I or -II did not increase VEGF concentrations, whereas the same levels of insulin tended to increase (p=0.02) VEGF at 72 hrs. In the presence of LH, again there was no effect of 1-10 ng/ml IGF-I or II at 2-3 days, but insulin significantly (p<0.001) increased VEGF concentrations during this interval. Notably, these results are markedly different from those we reported earlier using LGC from non-PCOS patients during IVF protocols (Fertil Steril Supp 3, 80: S278, abstract P477, 2003): (1) IGFs and insulin did not markedly alter DNA content during 3-day culture, and (2) at 1-10 ng/ml, both IGF-I and -II, but not insulin, increased VEGF levels in the culture media. Exposure to both insulin and IGFs, as well as LH, for greater than 24 hours promotes VEGF-A secretion by LGC from PCOS women. Thus, both endocrine and local factors play a role in regulating VEGF production by the ovulatory, luteinizing follicle. However, there appear to be differences in responses between cells from non-PCOS and PCOS women, with the latter more sensitive to insulin and less sensitive to IGFs in terms of granulosa cell VEGF-A production, particularly in the presence of LH. Insulin and IGFs also selectively promoted cell proliferation of granulosa cells from PCOS women. Whether these characteristics reflect inherent differences in the maturing follicle, or the responses from treatment during COS in women with PCOS remains to be established.